[CRIU] [PATCHv2 0/9] Large pages support for aarch64/ppc64

[Adrian Reber]

On Wed, Mar 28, 2018 at 05:19:44PM +0100, Dmitry Safonov wrote:
> v2 Changes:
> o Add patches for global variables in PIEs
> o Simplify ARCH_HAS_LONG_PAGES
> o cache PAGE_SIZE and PAGE_SHIFT values
>   (as nitpicked by Mike)
> 
> Adrian, I've removed your Tested-by, as the set differs a bit from
> the previous version. Mind to check that everything still works in v2?

Testing on s390 gives me this:

################### 3 TEST(S) FAILED (TOTAL 354/SKIPPED 33) ####################
 * zdtm/static/s390x_gs_threads(h)
 * zdtm/static/deleted_unix_sock(h)
 * zdtm/static/del_standalone_un(h)

My kernel has no guarded storage support, so that error is expected.

On ppc64le I get:

################### 5 TEST(S) FAILED (TOTAL 351/SKIPPED 33) ####################
 * zdtm/static/aio00(h)
 * zdtm/static/aio01(h)
 * zdtm/static/maps06(h)
 * zdtm/static/deleted_unix_sock(h)
 * zdtm/static/del_standalone_un(h)

The same test cases as above are also failing (deleted_unix_sock,
del_standalone_un) which probably should be investigated, but most of
the test cases are working.

On aarch64 all test cases are failing with errors like:

======================= Run zdtm/static/busyloop00 in h ========================
Start test
./busyloop00 --pidfile=busyloop00.pid --outfile=busyloop00.out
Run criu dump
=[log]=> dump/zdtm/static/busyloop00/36/1/dump.log
------------------------ grep Error ------------------------
(00.007988) PPB: 8 pages 4 segs 16 pipe 0 off
(00.008006) Sent msg to daemon 66 0 0
pie: 36: __fetched msg: 66 0 0
(00.008021) Wait for ack 66 on daemon socket
(00.008054) Error (criu/parasite-syscall.c:92): si_code=4 si_pid=36 si_status=11
(00.008064) Error (criu/parasite-syscall.c:100): 36 was stopped by 11 unexpectedly
------------------------ ERROR OVER ------------------------

################## 316 TEST(S) FAILED (TOTAL 351/SKIPPED 34) ###################

Not sure why it errors over, but the first few patches were working better.

I am not sure if it is the exact same hardware I am testing on, but the
same kernel.

		Adrian